Native Instruments Massive X Synth - Sequel to Massive (Out Now!)

[Urs]

And, AFAIK, harmonics added by saturation or distortion isn't aliasing.

That's correct and I think I would have noticed if he had assumed that. Hope you don't mind if I don't watch it again to verify - I have Massive X in my account and I'll be gone for a few hours

[Functional]

There's also a lot of confusion about his use of the aliasing term. This isn't really aliasing, is it? Not sure.

He used the term correctly. His concern was that, if oscillators were not properly bandlimited (which he did not conclude), aliasing would emphasize with audio rate modulation. That's one issue of concern, and one that a user manual could perfectly rule out. He furthermore demonstrates that aliasing does happen with filter distortion, as it does with pretty much every other synth out there. Hence, if there were two sources of aliasing, his concern is that there might be a "compound effect".

He does not draw any conclusions, he rather raises questions. A user manual could clarify.

I'm not sure about the point of these raised questions when we specify them to what we actually observed; so if you play a C10 (or something like that), there's some potentially significant aliasing, especially if you happen to play a _modulator_ at that range, which generally speaking don't have necessarily anti-aliasing measures especially when they're polyphonic to conserve CPU use.

So here's a question I'll raise: who cares? I don't play a C10 and if I use audio rate modulators, they will also not be going at those kind of rates. Is there really any other reason to do that except to take a piss?

I do agree that user manual would be great, but even user manual would not have saved us from this as the user manual probably won't explain that playing close to nyqvist frequency will produce relatively strong aliasing artifacts and that there probably aren't too many reasons to care about that when the nyqvist frequency in itself is at a range that is inaudible to most of us

Anyhoo, have fun with MX! Goes quite well with colourcopy & even twangström personal recommendation: try out Morse Chords patch, pay attention to the fifth modulator and try the patch with it off and also at mono. It's by far my favorite patch ever!

[Stefken]

Now if Massive X allowed me to do that with 4 WT oscs and 6 inserts and 2 filters, it'd be the synth of my dreams.

You could replicate that number by using two instances together, like layer mode on a pro 12 for example. No ideal I know but similar.

Yes, but then what you have is essentially just two different instances, which sadly doesn't help with expanding the routing of a single instance. Once you start using stuff like feedback, you'll quickly learn why two separate instances isn't even a workaround really.

Also another item on the wishlist: expanding the noise oscs altogether. I want to control their phase too. In fact, let us just load custom samples. Now that would be purrrrrrrfect. I've had so much fun with the noise oscs actually as it is, but especially the human-like and animal sounds are lacking as there is only essentially like a bird, two grizzly growls and two human voices. That's surprisingly a lot considering how much can be done with those already, but I want moar

I agree. I don't know why there is no way to insert a wavetable osc or a primary filter into one of the inserts as the code for these exists anyway, and for that matter why there are only three slots. If it is a CPU issue then its not like you don't have a choice.
Seems a strange design in some ways.

My take on it, is that they did it for performance and wanted to shortcircuit any discussions beforehand.

They already have a small riot on their hands regarding the avx case.
If people used 2 monarks which use a lot of juice, and got cpu issues, than they had to explain why. And then people would say: sure but Dune, Hive, ... can manage two. Why can't you? Yada yada yada.

So, my guess is, already having enough challenges, that they shortcircuited the whole discussion and just provided one main filter.

[Telematix]

He did switch filters on and off, and he did connect the oscillator directly to the output and it showed harmonics.

Yes, apparently one can filter those harmonics inside the (main) oscillator and then it looks clean.

The fact is that after those videos some users claim that this synth cannot produce pure sine.
But actually it can.

[vurt]

He did switch filters on and off, and he did connect the oscillator directly to the output and it showed harmonics.

Yes, apparently one can filter those harmonics inside the (main) oscillator and then it looks clean.

The fact is that after those videos some users claim that this synth cannot produce pure sine.
But actually it can.

i saw the sine

[Telematix]

[double post]

[Telematix]

[double post]

[Mushy Mushy]

He did switch filters on and off, and he did connect the oscillator directly to the output and it showed harmonics.

Yes, apparently one can filter those harmonics inside the (main) oscillator and then it looks clean.

The fact is that after those videos some users claim that this synth cannot produce pure sine.
But actually it can.

i saw the sine

And it opened up my mind. Ooooh I saw the sine.

[Symetry]

Modulating the Monark Filter via Performers seems to cause lots of audible clicks on each value jump. Is there a way to smooth this?

[vurt]

steve is stuck in a loop

[NatLife]

Testing it, interesting synthesizer so far. I'm not a fan of latest NI sounds, mostly i used only old Massive for the last 5-7 years in production, because of some plastic/ or deaf sounding in their production, before was in love with Reaktor 4/5, Absynth etc, i dunno, maybe it's because of NI changed their programming team or what else, or they thought that deaf/plastic sounding of their instruments is ok, but for such users like me is'nt, and i know enough people who also not used NI products for the last years, coz of this...but as for Massive X, it seems have same sound engine like all the latest NI synths, but! important, it can be very well boosted with a 64voices polyphony, +6 detune voices and 2 different stereo exciters..wow! plus alot alot of other interesting things, that need a more time for learn, so it seems a very very interesting synthesizer, not like others, yeah, but we no need like others) We already have Hive 2, Dune 3, Serum, that just sounds different from NI, so it can complete each other

[ghettosynth]

And, AFAIK, harmonics added by saturation or distortion isn't aliasing.

That's correct and I think I would have noticed if he had assumed that. Hope you don't mind if I don't watch it again to verify - I have Massive X in my account and I'll be gone for a few hours

At 19:48 he shows the distortion without aliasing and then states "all of the oscillators show these aliasing artifacts." It's reasonable for someone who speaks English to assume that "these" refers to the artifacts just shown, which are not aliasing artifacts.

At 21:04, after showing that the filter aliases, he refers to the effects as "compounding aliasing effects."

At 21:27 while showing simply distortion of the oscillators through the filter he claims that it's still "pretty bad," implying that filter saturation is aliasing. At no time does his language switch from the "compounding aliasing effects" to "saturation effects" and at no time past 21:27 does he demonstrate aliasing.

At 21:48 he asserts that he feels that this should be "neutral" here and that no such distortion, which is reasonable to conclude that he thinks is aliasing based on his own words, should exist when resonance is down and the filter is open. This of course suggests that he doesn't understand how saturation can be an integral part of the filter path and not necessarily only in the feedback path. Further, he makes no effort to quantify the impact of the saturation there, not that it matters, that's a completely different topic.

His claim is that the filter should be neutral here is, of course, a "conclusion." His claim that he "expects more from NI" further suggests that the behavior is incorrect.

Then at 22:40 he claims that it's totally fair to compare it to other modern synths that do the "whole aliasing thing properly." That is also a veiled conclusion based on no actual evidence.

Sorry mate, the faux humility combined with ignorance stinks to high heaven. At no time did he make any attempt to actually quantify the actual aliasing of the osc nor compare those levels to other synths.

[dionenoid]

steve is stuck in a loop

This whole thread is stuck in a loop

[Functional]

I agree. I don't know why there is no way to insert a wavetable osc or a primary filter into one of the inserts as the code for these exists anyway, and for that matter why there are only three slots. If it is a CPU issue then its not like you don't have a choice.
Seems a strange design in some ways.

Because wild speculation is so hot these days, let me participate in doing so:

They might have had a plan initially for A/B page for the top panel (everything before the VCA) while keeping the bottom panel the same as it is now. Due to rushed release, they didn't have the time to implement this and test it out properly (I'm assuming that it's not quite as easy to implement as just copypasteing everything).

There is no evidence to support this claim, but there is a single reason to believe it: the routing section is much bigger currently than it needs to be and the cable routing requires way too much precision too. Another mildly related reason might be the single filter aspect of it: that particular detail feels like a downgrade from Massive (even though overall this is a HUGE upgrade). It might very well be that they intended for it to have 2 filters initially, but the plan was for this A/B UI structure instead of having fair amount of clutter (like Massive had) or just pages and pages for different modules and stuff (i.e. Omnisphere).

Another mild reason I guess is that since there are only 3 inserts, it's kind of weird that third osc would really have to occupy one of them when all of them are extremely usable, including the utility tool. 3 of those poly inserts just isn't enough even if we don't take the oscs into account

But I wouldn't expect to see this anytime soon since we don't even have midi learn yet. And I must say - they did make the right call probably. What you can do with the current design is already amazing, there's no denying that. Doubling up the capacity surely would have expanded on this but I think there's other features that they should probably focus on more. Not to mention, they probably need more testing for doubling the capacity and how CPU's are going to handle it on the average, especially if there is no "draft mode".

[vurt]

steve is stuck in a loop

This whole thread is stuck in a loop

dont worry, once massivex is actually released it will move forward